Re: [Bioc-devel] Troubles using R/Bioc-devel with Mac OS X El Capitan 10.11.1

2015-11-18 Thread Ludwig Geistlinger
Just to mark that as solved: Seems as I've just been at the wrong place on the wrong time (by using R-devel r69640 downloaded yesterday). When compiling from the most recent R-devel_2015-11-17_r69648.tar.gz available from cran, I neither encountered the biocLite problem under linux nor the

Re: [Bioc-devel] A bioconductor package archive?

2015-11-18 Thread Hervé Pagès
Hi Gabe, On 11/18/2015 09:50 AM, Gabe Becker wrote: Hi all, Unless I'm missing something, there is no archive of prior source packages for Bioconductor, in the vein of CRAN's Web Archive. It has generally been accepted, I think, that if you need a version of a package - e.g., for precise

Re: [Bioc-devel] Unexpected behaviour with Assays and Vector classes

2015-11-18 Thread Hervé Pagès
Developers, On 11/16/2015 11:39 PM, Hervé Pagès wrote: Hi Aaron, On 11/16/2015 09:39 AM, Hervé Pagès wrote: [...] Anyway, you convinced me that we should not try to follow too closely the names model for how we treat metadata columns. If nobody objects, I'll change the behavior of the

[Bioc-devel] A bioconductor package archive?

2015-11-18 Thread Gabe Becker
Hi all, Unless I'm missing something, there is no archive of prior source packages for Bioconductor, in the vein of CRAN's Web Archive. It has generally been accepted, I think, that if you need a version of a package - e.g., for precise reproducibility - you can go to the still-existing release

[Bioc-devel] Troubles using R/Bioc-devel with Mac OS X El Capitan 10.11.1

2015-11-18 Thread Ludwig Geistlinger
Hi, I am experiencing troubles installing "Hmisc", especially its dependency "acepack", via > biocLite("acepack") in R-devel installed from the "R-devel-mavericks-signed.pkg" downloaded from https://r.research.att.com/. The error reads: BioC_mirror: https://bioconductor.org Using

Re: [Bioc-devel] Troubles using R/Bioc-devel with Mac OS X El Capitan 10.11.1

2015-11-18 Thread Ludwig Geistlinger
While I am following Jo's suggestions on installing gcc via homebrew on my mac laptop (is it normal that the "make bootstrap" command takes >20 min already?), I am in parallel trying to make R/Bioc-devel working on our institutional linux machines. I was able to configure/make R-devel here,

Re: [Bioc-devel] Troubles using R/Bioc-devel with Mac OS X El Capitan 10.11.1

2015-11-18 Thread Morgan, Martin
The suggestion on R-SIG-mac in the thread I cited is that homebrew is not the right way to go. This is from very established members of the R-community (Simon Urbanek, Brian Ripley), and I would not treat it lightly. Also, if you are installing R-devel (implied by use of Bioc 3.3), there are

Re: [Bioc-devel] Troubles using R/Bioc-devel with Mac OS X El Capitan 10.11.1

2015-11-18 Thread Henrik Bengtsson
For your Linux problem: You might have build R differently from what you had in the past, e.g. static vs dynamic. Try to explicitly re-install those failed dependences and see if you any further. If so, you might wanna remove all your installed packages and start over with a fresh package

Re: [Bioc-devel] Troubles using R/Bioc-devel with Mac OS X El Capitan 10.11.1

2015-11-18 Thread Rainer Johannes
Could be that gcc via homebrew takes a while… especially if it is compiling that stuff from source… > On 18 Nov 2015, at 14:46, Ludwig Geistlinger > wrote: > > While I am following Jo's suggestions on installing gcc via homebrew on my > mac laptop (is it